Unlock instant, AI-driven research and patent intelligence for your innovation.

Learning lighting from various portraits

A lighting source, image technology, applied in image communication, image data processing, 3D image processing and other directions

Pending Publication Date: 2022-08-02
GOOGLE LLC
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The problem in both still photo and video applications is matching the lighting of a real world scene so that the rendered virtual content plausibly matches the appearance of the scene

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Learning lighting from various portraits
  • Learning lighting from various portraits
  • Learning lighting from various portraits

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0098] Example 1: A method that includes:

[0099] receiving image training data representing a plurality of images, each of the plurality of images including at least one of a plurality of faces, each of the plurality of faces having been combined as described by a physical or virtual environment formed from an image of one or more faces illuminated by at least one of a plurality of illumination sources, each of the plurality of illumination sources already in a plurality of the corresponding one of the orientations; and

[0100] A prediction engine is generated based on the plurality of images, the prediction engine configured to generate a predicted illumination profile from input image data representing an input human face.

example 2

[0101] Example 2: The method of Example 1, further comprising:

[0102] combining the images of the one or more faces illuminated by the at least one illumination source of the plurality of illumination sources to synthetically render each face of the plurality of faces to appear as powered by high dynamic range HDR Illuminated by the lighting environment.

example 3

[0103] Example 3: The method of example 2, wherein combining the images comprises:

[0104] The HDR lighting environment is generated based on low dynamic range LDR images of a set of reference objects, each reference object in the set of reference objects having a corresponding bidirectional reflectance distribution function BRDF.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

Techniques for estimating illumination from a portrait include generating an illumination estimate from a single image of a face based on a machine learning (ML) system using multiple bidirectional reflectance distribution functions (BRDFs) as a loss function. In some embodiments, the ML system is trained using images of the face formed with HDR illumination calculated from LDR images. Technical schemes include training an illumination estimation model in a supervised manner using a portrait and its corresponding ground real illumination data set.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS [0002] This application is a non-provisional application to US Provisional Patent Application No. 62 / 704,657, filed May 20, 2020, entitled "LEARNING ILLUMINATION FROMPORTRAITS," the disclosure of which is incorporated by reference The whole is incorporated here. technical field [0003] The present disclosure relates to determining lighting from portraits for use in, for example, augmented reality applications. Background technique [0004] A problem in both still photo and video applications is matching the lighting of the real world scene so that the rendered virtual content plausibly matches the appearance of the scene. For example, lighting schemes can be designed for augmented reality (AR) use cases with world-facing cameras, such as in the rear camera of a mobile device, where one might want to render a synthetic object (such as a piece of furniture) into a real-world scene Live camera feed. SUMMARY OF THE INVENTION ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06T15/50G06V10/60G06V10/774H04N5/235H04N5/232
CPCG06T15/506G06V10/60H04N23/611H04N23/741G06F18/214
Inventor 克洛伊·勒让德尔保罗·德伯韦克马万钧罗希特·潘迪肖恩·瑞安·弗朗切斯科·法内洛克里斯蒂娜·彤
Owner GOOGLE LLC